Texture pattern image generation by regular Markov chain

Abstract

A method of graytone texture pattern generation using a regular Markov chain is presented. The procedure arranges the generated gray tones in a sequence along a scan line. The transition probability matrix of the Markov chain directly determines the spatial co-occurrence probabilities of gray tones in the generated image. The generated image can be rotated and arithmetically combined to produce images of additional texture patterns. The paper illustrates the variety of texture which can be produced by this method.